2D_Simple_Channel_Flow

[C0st4s]

Hello guys,

I am new user to Code_Saturne and I have set a simple channel flow case with periodic inlet/outlet. The dimensions of my geometry is 1m along x-direction (periodic), 1m along z-direction (symmetry) and 2m along y-direction (top and bottom walls). I have generated a hexahedral mesh with 400 faces along y-direction , 1 cell in z-direction and 5 cells along x-direction. I have run the case with k-epsilon model for steady state and I am using a source term for the pressure gradient of the inlet/outlet. The Retau is 10000 (Re approx 5.6e5),tauwall=1, rho=1, with mu=0.0001 and initialising velocity along x-direction around 28m/s. I have two questions:
1)When I plot a profile along a line (y-direction) of U+ vs Y+ my line begins on 30 (Yplus=30). If I refine the mesh can this line begin from 1 let's say?
2)If I want to solve with different models like Reynolds Stress Models or k-omega or v2-f model do I need to create a new mesh for each simulation or shall I use the same?

If you could help, that would be great!

Thanks
Constantinos

[Yvan Fournier]

Hello,

1) Yes

2) You can use the same mesh, but some models are better adapted to more refined meshes (low-Reynolds) near the boundary, other to coarser meshes. The best practice recommendations in the documentation section of the main web site provide some recommendations.
In most cases, 2 meshes should be enough (for Low-Reynolds, models, up to y+=1, the other for higher Reynolds models, with y+ around 30 near the boundary). Some models also allow for scalable options, but having the correct refinement is safer.

Regards,

Yvan

[C0st4s]

Yvan,

Many thanks.

Constantinos

[C0st4s]

Hi,

A few more questions related to the same 2D_channel flow case.
1)What subroutine shall I use if I want to impose uniform heat flux on wall boundaries for my channel?

2)Can this be done via GUI?

3) How can I print the value of Tplus for each cell along a y-direction line?

Kind Regards,
Constantinos

[Yvan Fournier]

Hello,

For 1) and 2), did you try the GUI ?

For 3), you can obtain T+ using addition settings for the boundary faces (check cs_user_parameters.f90 example), but obtaining it for cells is more complex. Also, do you want to print it to a file or postprocess it another way ?

Regards,

Yvan

[C0st4s]

Hello Yvan,

Thanks again for your response. Yes I have found how to impose heat flux in GUI but I want to extract the Tplus along a line (i.e cells and boundary ) by printing the results in a different file (i.e csv)

Cheers,
Constantinos

[Yvan Fournier]

Hello,

For T+, check how to activate it in the cs_user_parameters.f90 examples.

To output it along a line near the boundary, check the examples for probes and profiles at the end of this page: https://www.code-saturne.org/cms/sites/ ... ocess.html.

You'll need to adapt the code, but the essentials are there, for T+ along the boundary. Inside the volume, I think you will need more involved coding.

Regards,

Yvan